钻井岩石力学参数三维建模方法及其现场应用

THREE-DIMENSIONAL MODELING METHOD FOR DRILLING ROCK MECHANICS AND ITS FIELD APPLICATION

在地层空间中建立岩石力学参数三维数据模型可指导钻井设计与施工,但常规建模方法的精度和分辨率有限。考虑到层速度是求取钻井岩石力学参数的基础数据,理论分析发现地层波速与地震属性之间存在着较复杂的映射关系,而神经网络学习算法具备识别这种定量关系的能力,基于识别结果可预测三维地震层速度,据此提出了钻井岩石力学参数三维建模新方法。在提取得到目标工区内完钻井的声波测井和井旁地震属性的基础上,将各套地层中的声波速度和对应的地震属性作为学习样本对,使用小波神经网络分层识别波速和地震属性之间的非线性函数关系,再将这种关系延拓至工区三维空间上,利用地震信息依次建立地震层速度和各类钻井岩石力学参数三维模型,并将其用于指导钻井工程。本方法在鄂尔多斯南部YS地区进行了应用,地层三压力和岩石可钻性三维建模成果具有较高的精确度和分辨率,依据模型进行的钻井工艺技术优化取得了良好的提高钻速、减少复杂的效果。

The three-dimensional models of rock mechanics established in the formation space can guide the drilling planning and drilling operations,but the precision and resolution of the conventional modeling method are limited.Theoretical analysis shows that there is a more complex mapping relationship between formation wave velocity and seismic attributes,and neural network learning algorithm is able to identify this quantitative relationship.Considering the interval velocity as the basic data for obtaining the drilling rock mechanical parameters,the 3D interval velocity can be predicted based on the learning results of neural network,therefore a new method for 3D modeling of drilling rock mechanics is proposed.On the basis of extracting acoustic logging and borehole seismic attributes in the target area,the acoustic velocity and corresponding seismic attributes in various strata are taken as learning samples.The wavelet neural network is used to distinguish the nonlinear function relation between the wave velocity and the seismic attribute,and then the relationship is extended to the three dimensional space of the target area.According to the method above,the seismic interval velocity and various kinds of the drilling rock mechanical parameters are established in turn by using seismic information,and modeling results are used to guide the drilling engineering.This method has been applied in the YS area of southern Ordos Basin and three dimensional modeling results of three formation pressures and rock drillability are of high accuracy and high resolution.Further optimization of the drilling technology according to the modeling has achieved good effects in improving drilling speed and reducing drilling complications.